I have made an attempt at a solution, https://mudg.fly.dev. I am a solo-dev, and would be interested in feedback/feature requests.

Completely agree. I feel no pressure to constantly upgrade my Elixir versions. I just look at the changes and there is often useful features that make me want to upgrade, as opposed the feeling of dread when I am pushed to upgrade.

Really interesting post. I ran into some of the limitations of working with tables and LLM's last year.

I experimented with an approach to use the llm to generate a bespoke transformation machine that uses an LLM to generate a series of transform steps to extracting key data from large data sets.

https://tombers.github.io/oblique-angles/ai/education/2025/0...

What is a good way to test for better than human intelligence?

Proposal - The Skibidi toilet test

So the test is very simple:

Could an artificial intelligence create a piece of media on demand that reached similar levels of cultural impact as Skibidi toilet?

Why?

1) This is not currently humanly possible. Many try, and some are better than others, but it is fundamentally impossible to predict.

2) Out of distribution By definition it has to be something unique, eliciting a thats cool response.

3)Models other minds The tone \ content \ semiotics has to speak to a large enough group of people and worth sharing.

4) Surfing the Geist To both understand and respond to the Geist, to directly observe the "anima mundi, spiritus universi"

One aspect of using LLM's to code that I have not seen mentioned is the "loss of attachment" to code in my projects.

For example, working on my project (https://mudg.fly.dev) I wanted to experiment with a new FE graph library. I asked an llm the fantastic (https://tidewave.ai) and it completely implemented a solution, which turned out to be worse than the current solution.

If I had spent many hours working on a solution I might have been more "attached" to the solution and kept with it.

Perhaps this is a good thing? (https://en.wikipedia.org/wiki/Nonattachment_(philosophy))

Have you considered that LLMs are also biased against new languages and libraries, so the code quality will be worse compared to something more established regardless of what you personally think/feel?

I have been using Tidewave since it released, and use it for a relatively complex Phoenix app (https://github.com/TomBers/dialectic).

The ability to have access to the context of the code, run tests and edit the code in place, are capabilities I could not even dream about when I started coding.

How Jose manages to see the potential of a technology (Erlang, LLM's) and produce something so elegant and useful is amazing. It has certainly changed my life.